GPS World, April 2017

UAV FIGURE 2 LTE module of the MATRIX SDR and corresponding LabVIEW VIs FIGURE 3 Tightly coupled cellular aided INS framework conventional low complexity DLLs Several methods can be used to estimate the channel parameters including the TOA multiple signal classification MUSIC estimation of signal parameters via rotational invariance techniques ESPRIT and space alternating generalized expectationmaximization SAGE algorithms LTE Receiver Structure The LTE navigation receiver exploits SSS PSS and CRS and consists of four stages Acquisition In this step the received signal is correlated with the locally generated PSS and SSS signals to obtain the frame start time estimate Doppler frequency estimate and the eNodeBs cell ID System information extraction In LTE systems the bandwidth can be assigned to different values The actual value of the bandwidth is provided to the UE by the eNodeB in a block called master information block MIB When user equipment enters an LTE network it starts receiving signals with the lowest possible bandwidth After obtaining the frame start time it is possible to convert the LTE signals into frame structure by executing the steps discussed in the LTE Frame Structure section in reverse order Then the UE decodes the MIB and obtains the actual bandwidth The UE can then increase the sampling rate to as high as the signal bandwidth Due to the near far effect on the PSS signal it is not possible to acquire all the available eNodeBs in the environment Each eNodeB provides the list of its neighboring cell IDs to the UE in the system information block SIB After obtaining the frame start time and the actual transmission bandwidth the UE can decode the SIB to obtain the neighboring cell IDs Tracking The receiver starts tracking the SSS using components of the tracking loop a frequency locked loop FLL assisted PLL to track the carrier phase and a carrieraided DLL to track the code phase Timing information extraction To overcome the error due to multipath in tracking the SSS the CRS is used For this purpose by knowing the CRS sequence and the received signal the channel frequency response is first estimated Then the channel impulse response is obtained by taking an IFFT of the channel frequency response Finally the first 22 GPS WORLD WWW GPSWORLD COM APRIL 2017 peak of the channel impulse response is detected which represents the line of sight TOA FIGURE 2 illustrates the block diagram of the LTE module of the MATRIX SDR and the corresponding LabVIEW VIs CELLULAR AIDED INERTIAL NAVIGATION To correct INS errors using cellular pseudoranges an extended Kalman filter EKF framework similar to a traditional tightly coupled GNSS aided INS integration strategy is adopted with the added complexity that the cellular BTSs states position and clock error states are simultaneously estimated alongside the navigating vehicles states position velocity attitude IMU measurement error states and receiver clock error states This framework is composed of two modes Mapping Mode The EKF produces estimates and associated estimation error covariances of both the navigating vehicle and the cellular BTSs states augmented in x using both GNSS SV and cellular BTS pseudoranges Between aiding corrections the EKF produces the state prediction x and prediction error covariance P using INS model and receiver and cellular BTS clocks models When an aiding source is available either a GNSS SV or cellular BTS pseudorange the EKF produces a state estimate update x and associated estimation error covariance P SLAM Mode The cellular aided INS framework enters a SLAM mode when GNSS pseudoranges become unavailable In this mode INS errors are corrected using cellular BTS pseudoranges and the cellular BTSs state estimates provided from the mapping mode As the autonomous vehicle navigates it simultaneously continues to refine the BTSs state estimates FIGURE 3 illustrates a high level diagram of the cellular aided INS framework SIMULATION RESULTS To evaluate the performance of this cellular aided INS framework presented simulations were conducted of a UAV equipped with the MATRIX SDR navigating in downtown Los Angeles while exploiting ambient cellular signals Two navigation systems were employed to estimate the trajectory of the UAV a traditional tightly coupled GPS aided INS